A marine lithium battery starting power supply BMS protection system

Abstract

The present invention relates to the technical field of electric ships, and discloses a marine lithium battery starting power supply BMS protection system, the system comprises: a data extraction module, a state calculation module, a fault detection module, and an automatic adjustment module, wherein: the data extraction module is used to collect key parameter data of the marine lithium battery starting power supply, and extract features of the key parameter data to obtain extracted features; the state calculation module is used to calculate the battery state of the marine lithium battery starting power supply using the extracted features; the fault detection module generates a dynamic balancing management strategy for the marine lithium battery starting power supply according to the fault location, and applies the dynamic balancing management strategy to the marine lithium battery starting power supply BMS. The present invention can solve the problem of low safety of the existing marine lithium battery starting power supply.

Description

Technical Field

[0001] The present invention relates to the technical field of electric ships, and in particular to a marine lithium battery starting power supply BMS protection system. Background Art

[0002] With the continuous advancement of electric ship technology, marine lithium batteries have become one of the key power sources to promote this change. It not only determines the operating efficiency and economy of the ship, but also directly affects the safety and environmental protection of navigation. However, the application of marine lithium batteries also faces a series of challenges, especially in safety management such as real-time monitoring and status assessment.

[0003] Traditional monitoring methods are difficult to adapt to the complex and changeable marine environment, making it difficult to achieve all-round, high-precision real-time monitoring of battery status; at the same time, most battery management systems use a relatively simple status assessment method, ignoring the influence of important factors such as temperature and internal resistance, affecting the accuracy of key indicators such as battery health status (SOH) and charge status (SOC). In addition, the existing fault diagnosis technology is mostly post-processing, lacking an effective early warning mechanism, increasing safety risks and maintenance costs. In summary, the existing marine lithium battery starting power supply BMS protection system still has the problem of low safety of marine lithium battery starting power supply. Summary of the invention

[0004] The invention provides a marine lithium battery starting power supply BMS protection system, the main purpose of which is to solve the problem of low safety of the existing marine lithium battery starting power supply.

[0005] To achieve the above-mentioned purpose, the present invention provides a marine lithium battery starting power supply BMS protection system, including: a data extraction module, a state calculation module, a fault detection module, and an automatic adjustment module, specifically:

[0006] A data extraction module is used to collect key parameter data of a marine lithium battery starting power supply, perform feature extraction on the key parameter data, and obtain extracted features;

[0007] A state calculation module, used to calculate the battery state of the marine lithium battery starting power supply using the extracted features;

[0008] A fault detection module, used to perform potential fault detection on the marine lithium battery starting power supply by using the battery status and the extracted features to obtain a fault location;

[0009] The automatic adjustment module is used to generate a dynamic balancing management strategy for the marine lithium battery starting power supply according to the fault location, and apply the dynamic balancing management strategy to the marine lithium battery starting power supply BMS to achieve dynamic management of the marine lithium battery starting power supply.

[0010] Optionally, when the data extraction module performs the function of extracting features from the key parameter data to obtain extracted features, it is specifically used to:

[0011] Performing data filtering on the key parameter data to obtain a filtering result;

[0012] Standardizing the filtering result to obtain standardized data;

[0013] The standardized data is subjected to Fourier transform to obtain frequency domain features, and Fourier transform is performed using the following formula:

[0014] ;

[0015] in, is the frequency domain feature frequency components, For the Standardized data, is an imaginary unit, is the total number of standardized data, is the frequency index, ranging from 0 to , is the preset value of pi, is the base of the preset natural logarithm;

[0016] Mutual information screening is performed on the frequency domain features to obtain extracted features.

[0017] Optionally, when the data extraction module performs the function of performing mutual information screening on the frequency domain features to obtain extracted features, it is specifically used to:

[0018] Calculate the mutual information value between each frequency domain feature and the preset target variable, and use the following formula to calculate the mutual information value:

[0019] ;

[0020] in, is the mutual information value, For the frequency domain features, is the target variable, For the Frequency domain features The value of And the target variable is The probability of For the Frequency domain features The value of Probability, The value of the target variable is The probability of For the Frequency domain features A specific value of Represents the target variable A specific value of ;

[0021] Sorting the frequency domain features using the mutual information values ​​to obtain a sorting result;

[0022] According to the sorting results, a preset number of frequency domain features are sequentially screened out as extracted features.

[0023] Optionally, when the state calculation module performs the function of calculating the battery state of the marine lithium battery starting power supply using the extracted features, it is specifically used to:

[0024] Reshaping the extracted features to obtain a three-dimensional array;

[0025] Perform forget gate calculations on the three-dimensional array in sequence to obtain a forget output;

[0026] Performing input gate calculation on the three-dimensional array to obtain an input value;

[0027] Calculate candidate information according to the input value;

[0028] Calculate the cell state according to the candidate information, the input value and the forgotten output;

[0029] Calculating output information according to the cell state;

[0030] The battery state is calculated based on the output information and the cell state.

[0031] Optionally, when the state calculation module performs the function of calculating the cell state according to the candidate information, the input value and the forgotten output, it is specifically used to:

[0032] Multiply the input value and the candidate information element by element to obtain current information;

[0033] Adding the forgotten output and the current information to obtain a state value of the cell to be processed;

[0034] Performing numerical mapping on the state value of the cell to be processed to obtain the cell state.

[0035] Optionally, when the fault detection module performs the function of performing potential fault detection on the marine lithium battery starting power supply by using the battery status and the extracted features to obtain the fault location, it is specifically used to:

[0036] Performing feature fusion on the battery state and the extracted features to obtain a feature vector;

[0037] Performing category confidence calculation on the feature vector to obtain confidence;

[0038] converting the confidence into a probability distribution;

[0039] The type and location of the potential fault are determined according to the probability distribution.

[0040] Optionally, when the automatic adjustment module executes the function of generating a dynamic balancing management strategy for the marine lithium battery starting power supply according to the fault location, it is specifically used to:

[0041] Calculating the performance difference between batteries according to the fault location and the battery status;

[0042] generating a balance control parameter according to the performance difference;

[0043] The balance control parameters of the marine lithium battery starting power supply are adjusted to obtain a dynamic balance management strategy.

[0044] Optionally, when the automatic adjustment module performs the function of calculating the performance difference between batteries according to the fault location and the battery status, it is specifically used to:

[0045] Performing a performance status comparison on the battery status to obtain a comparison result;

[0046] Perform fault correlation according to the comparison result and the fault location to obtain a correlation result;

[0047] The correlation result and the comparison result are used to quantify the difference of the marine lithium battery starting power supply to obtain the performance difference.

[0048] Optionally, when the automatic adjustment module performs the function of adjusting the balance control parameters of the marine lithium battery starting power supply to obtain a dynamic balance management strategy, it is specifically used to:

[0049] generating a preliminary update scheme according to the balance control parameters;

[0050] Performing fitness function evaluation on the preliminary updating scheme to obtain a fitness value;

[0051] The preliminary update scheme is iterated based on the fitness value to obtain a dynamic balance management strategy.

[0052] Optionally, when the automatic adjustment module applies the dynamic balancing management strategy to the marine lithium battery starting power supply BMS to realize the function of dynamic management of the marine lithium battery starting power supply, it is specifically used to:

[0053] Transmit the dynamic balancing management strategy to the BMS system through the communication interface;

[0054] The BMS system receives the dynamic balancing management strategy and parses the balancing control parameters and control instructions in the dynamic balancing management strategy;

[0055] Based on the control instruction, the control parameters in the BMS system are adjusted according to the balance control parameters in the dynamic balance management strategy.

[0056] The present invention finds battery cells with abnormalities (such as voltage, temperature or current not within the normal range) and combines the current state of the battery (such as charging level, discharge rate) to calculate the impact of these abnormalities. The system can judge the performance difference of the fault on the entire battery pack, thereby determining which batteries need higher priority processing, so that each battery cell operates under the most suitable conditions to avoid problems such as overheating and over-discharging of a single battery; by using a balancing algorithm to ensure that all battery cells operate stably within a safe range and dynamically adapt to changes in battery performance, the battery life is extended and safety is improved. Therefore, the marine lithium battery starting power supply BMS protection system proposed by the present invention can solve the problem of low safety of existing marine lithium battery starting power supplies. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A functional module diagram of a marine lithium battery starting power supply BMS protection system provided by one embodiment of the present invention;

[0058] Figure 2 A schematic diagram of a mutual information screening process provided by an embodiment of the present invention;

[0059] Figure 3 A schematic diagram of a process for generating a dynamic balancing management strategy provided by an embodiment of the present invention;

[0060] Figure 4 A schematic flow chart of a BMS protection method for a marine lithium battery starting power supply provided by one embodiment of the present invention.

[0061] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0062] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0063] Reference Figure 1 As shown, it is a functional module diagram of a marine lithium battery starting power supply BMS protection system provided by one embodiment of the present invention. In this embodiment, the marine lithium battery starting power supply BMS protection system 100 can be installed in an electronic device. According to the functions implemented, the marine lithium battery starting power supply BMS protection system 100 may include a data extraction module 101, a state calculation module 102, a fault detection module 103, and an automatic adjustment module 104. The module described in the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0064] In the embodiment of the present invention, the data extraction module 101 includes collecting key parameter data of a marine lithium battery starting power supply, performing feature extraction on the key parameter data, and obtaining extracted features;

[0065] In the embodiment of the present invention, the state calculation module 102 comprises calculating the battery state of the marine lithium battery starting power supply using the extracted features;

[0066] In the embodiment of the present invention, the fault detection module 103 includes performing potential fault detection on the marine lithium battery starting power supply using the battery status and the extracted features to obtain the fault location;

[0067] In an embodiment of the present invention, the automatic adjustment module 104 includes generating a dynamic balancing management strategy for the marine lithium battery starting power supply according to the fault location, applying the dynamic balancing management strategy to the marine lithium battery starting power supply BMS, and realizing dynamic management of the marine lithium battery starting power supply.

[0068] In detail, the modules described in the marine lithium battery starting power supply BMS protection system 100 described in the embodiment of the present invention adopt the same technical means as the marine lithium battery starting power supply BMS protection system described in the accompanying drawings when in use, and can produce the same technical effects, which will not be repeated here.

[0069] In the following, in combination with specific embodiments, each component and specific working process of the marine lithium battery starting power supply BMS protection system are described respectively:

[0070] The data extraction module 101 is used to collect key parameter data of the marine lithium battery starting power supply, perform feature extraction on the key parameter data, and obtain extracted features.

[0071] In the embodiment of the present invention, by collecting key parameter data of the battery (such as voltage, temperature, charging efficiency, internal resistance, etc.), the health state (SOH) of the battery can be evaluated in real time. By extracting features, the working state of the battery can be more comprehensively understood, such as whether the battery is in a decay stage or whether there is a potential fault.

[0072] In the embodiment of the present invention, the key parameter data of the marine lithium battery starting power supply refers to various parameters of the battery pack collected in real time by connecting various sensors or a battery management system (BMS), such as temperature, current, voltage, internal resistance, etc.

[0073] In the embodiment of the present invention, when the data extraction module performs the function of extracting features from the key parameter data to obtain the extracted features, it is specifically used to:

[0074] Performing data filtering on the key parameter data to obtain a filtering result;

[0075] Standardizing the filtering result to obtain standardized data;

[0076] Performing Fourier transform on the standardized data to obtain frequency domain features;

[0077] Mutual information screening is performed on the frequency domain features to obtain extracted features.

[0078] In the embodiment of the present invention, the filter signals preset by the key parameter data are multiplied to obtain a smoothed signal.

[0079] In an embodiment of the present invention, for each key parameter data, its mean value and standard deviation in the entire data set are first calculated, and then each key parameter data is standardized, that is, the mean value is subtracted from each key parameter data, and the result is divided by the standard deviation to obtain standardized data.

[0080] In the embodiment of the present invention, the following formula may be used to perform Fourier transform:

[0081] ;

[0082] in, is the frequency domain feature frequency components, For the Standardized data, is an imaginary unit, is the total number of standardized data, is the frequency index, ranging from 0 to , is the preset value of pi, The base of the default natural logarithm.

[0083] Ginseng Figure 2 In the embodiment of the present invention, when the data extraction module performs the function of performing mutual information screening on the frequency domain features to obtain extracted features, it is specifically used to:

[0084] S21, calculating the mutual information value between each frequency domain feature and the preset target variable;

[0085] S22, sorting the frequency domain features using the mutual information values ​​to obtain a sorting result;

[0086] S23. According to the sorting results, a preset number of frequency domain features are sequentially selected as extraction features.

[0087] In detail, the target variable refers to the goal that is desired to be achieved through feature extraction, such as predicting battery status and identifying fault types.

[0088] In detail, the mutual information value is calculated using the following formula:

[0089] ;

[0090] in, is the mutual information value, For the frequency domain features, is the target variable, For the Frequency domain features The value of And the target variable value is The probability of For the Frequency domain features The value of Probability, The value of the target variable is The probability of For the Frequency domain features A specific value of Represents the target variable A specific value of .

[0091] In the embodiment of the present invention, the extracted features can help monitor whether the battery has abnormalities during operation, such as overheating, voltage instability, or increased internal resistance, etc. By calculating the battery status in real time, potential problems can be discovered in advance and intervention can be made to prevent system failures.

[0092] The state calculation module 102 is used to calculate the battery state of the marine lithium battery starting power supply using the extracted features.

[0093] In the embodiment of the present invention, by extracting the characteristics of the battery, the health status of the battery can be evaluated in real time. Timely understanding the performance changes of the battery helps to determine whether the battery is in a good working condition and warn of possible faults.

[0094] In an embodiment of the present invention, when the state calculation module performs the function of calculating the battery state of the marine lithium battery starting power supply using the extracted features, it is specifically used to:

[0095] Reshaping the extracted features to obtain a three-dimensional array;

[0096] Perform forget gate calculations on the three-dimensional array in sequence to obtain a forget output;

[0097] Performing input gate calculation on the three-dimensional array to obtain an input value;

[0098] Calculate candidate information according to the input value;

[0099] Calculate the cell state according to the candidate information, the input value and the forgotten output;

[0100] Calculating output information according to the cell state;

[0101] The battery state is calculated based on the output information and the cell state.

[0102] In the embodiment of the present invention, data reshaping of the extracted features refers to arranging the extracted features in chronological order and representing them in the form of a three-dimensional array, i.e., a three-dimensional array; the shape of the three-dimensional array is usually (number of samples, time step, number of features). A sample refers to a time series composed of key parameters collected at each time point, a time step represents the number of time points in the time series, and a number of features represents the number of extracted features at each time point.

[0103] In the embodiment of the present invention, forget gate calculations are performed on the three-dimensional array in sequence through a neural network layer or matrix operation, and a value between 0 and 1 is output. The forget output determines how much old information will be retained.

[0104] In an embodiment of the present invention, the input gate calculation for the three-dimensional array is performed through matrix multiplication and activation function (such as Sigmoid function), and a vector is output. The input value determines which information is updated.

[0105] In the embodiment of the present invention, the candidate information is calculated by a neural network layer (usually a nonlinear activation function, such as tanh), and by performing information transformation on the input information, the information can be flexibly selected and updated, thereby better processing complex sequence data.

[0106] In the embodiment of the present invention, the cell state refers to a vector that can store long-term information about the sequence at each time step.

[0107] In the embodiment of the present invention, when the state calculation module performs the function of calculating the cell state according to the candidate information, the input value and the forgotten output, it is specifically used to:

[0108] Multiply the input value and the candidate information element by element to obtain current information;

[0109] Adding the forgotten output and the current information to obtain a state value of the cell to be processed;

[0110] Performing numerical mapping on the state value of the cell to be processed to obtain the cell state.

[0111] In detail, the input value and the candidate information are both vectors. By multiplying the input value and the candidate information according to the element position in the vector, if an element in the input value is close to 1, it means that the element in the corresponding candidate information will be completely retained and added to the cell state, and the final result is the current information.

[0112] Specifically, the forgotten output is multiplied element by element with a preset historical information value vector to obtain old information, and then the old information is added to the current information to obtain the state value of the cell to be processed.

[0113] In detail, the numerical mapping is accomplished through an activation function, such as a hyperbolic tangent function (tanh), and the final cell state is obtained by substituting the cell state value to be processed into the activation function, thereby ensuring that the value of the cell state is within the range of [-1,1].

[0114] In an embodiment of the present invention, the output information calculated based on the cell state is used to calculate the input of the next layer or the output of the final state, and a linear combination is obtained by performing a nonlinear transformation on the cell state. The result of the linear combination is transformed by an activation function to obtain the output of the output gate, that is, the output information, which is used to represent the main information of the current time step.

[0115] In the embodiment of the present invention, the battery state is calculated according to the output information and the cell state by performing element-by-element multiplication of the output information and the cell state.

[0116] In detail, the battery status is evaluated and predicted through a "neural network", where the "forget gate", "input gate", "cell status" and other steps are all simulating the brain's memory and processing mechanism to accurately extract and combine historical information with current data, and finally calculate the battery's health status. These steps help the system find patterns in the multi-dimensional characteristics of the battery (such as voltage, current, temperature, etc.), thereby identifying the battery status in real time or predictively.

[0117] In the embodiment of the present invention, the status score or health status of the battery is calculated by combining the output information and the cell status, and the output is a vector or scalar representing the battery status for final fault detection or health assessment.

[0118] In the embodiment of the present invention, the extracted features can help monitor whether the battery has abnormalities during operation, such as overheating, voltage instability or increased internal resistance. By calculating the battery status in real time, potential problems can be discovered in advance and intervention can be made to prevent system failures.

[0119] The fault detection module 103 is used to perform potential fault detection on the marine lithium battery starting power supply by using the battery status and the extracted features to obtain the fault location.

[0120] In an embodiment of the present invention, by real-time monitoring of the status and characteristics of the battery, potential problems can be detected before a fault occurs. This predictive fault detection can provide a more adequate response time for the marine lithium battery starting power supply and avoid damage to the battery and the system caused by sudden faults.

[0121] In an embodiment of the present invention, when the fault detection module performs the function of performing potential fault detection on the marine lithium battery starting power supply by using the battery status and the extracted features to obtain the fault location, it is specifically used to:

[0122] Performing feature fusion on the battery state and the extracted features to obtain a feature vector;

[0123] Performing category confidence calculation on the feature vector to obtain confidence;

[0124] converting the confidence into a probability distribution;

[0125] The type and location of the potential fault are determined according to the probability distribution.

[0126] In the embodiment of the present invention, the battery status data and the extracted feature data are concatenated into a feature vector.

[0127] In the embodiment of the present invention, a support vector machine (SVM) or other classification models are used to calculate the category confidence corresponding to the feature vector.

[0128] In the embodiment of the present invention, the confidence is converted into a probability distribution through the Softmax function, so that the confidence of each fault category is in the interval [0,1], which is convenient for determining the specific fault type and location.

[0129] In the embodiment of the present invention, the type and specific location of the fault are automatically determined according to the fault probability distribution, and the fault type and fault location are automatically generated by writing a rule engine or a conditional logic program to analyze the probability distribution.

[0130] In the embodiment of the present invention, through fault detection, the system can identify which batteries have potential performance problems (such as decreased battery capacity, increased internal resistance, etc.), and isolate them or reduce their load to avoid excessive use. This can reduce excessive loss of unhealthy batteries, thereby extending the service life of the entire battery pack.

[0131] The automatic adjustment module 104 is used to generate a dynamic balancing management strategy for the marine lithium battery starting power supply according to the fault location, and apply the dynamic balancing management strategy to the marine lithium battery starting power supply BMS to achieve dynamic management of the marine lithium battery starting power supply.

[0132] In the embodiment of the present invention, by adjusting the battery charging and discharging strategy according to the fault location, it is possible to effectively avoid over-discharging or over-charging of certain batteries due to faults, improve the safety of the marine lithium battery starting power supply, and extend the overall service life of the battery pack.

[0133] See also Figure 3 In the embodiment of the present invention, when the automatic adjustment module performs the function of generating a dynamic balancing management strategy for the marine lithium battery starting power supply according to the fault location, it is specifically used to:

[0134] S31, calculating the performance difference between batteries according to the fault location and the battery status;

[0135] S32, generating a balance control parameter according to the performance difference;

[0136] S33, adjusting the balance control parameters of the marine lithium battery starting power supply to obtain a dynamic balance management strategy.

[0137] In the embodiment of the present invention, when the automatic adjustment module performs the function of calculating the performance difference between batteries according to the fault location and the battery status, it is specifically used to:

[0138] Performing a performance status comparison on the battery status to obtain a comparison result;

[0139] Perform fault correlation according to the comparison result and the fault location to obtain a correlation result;

[0140] The correlation result and the comparison result are used to quantify the difference of the marine lithium battery starting power supply to obtain the performance difference.

[0141] In detail, technical means such as standard deviation, range, weighted average method, etc. can be used to compare data and obtain performance status comparison results of each battery.

[0142] In detail, by analyzing the impact of different fault types on various performance indicators (such as SOC, charging efficiency, internal resistance, etc.), a correlation matrix between fault types and performance indicators is constructed, and then the Pearson correlation coefficient is used to measure the linear correlation between fault types and performance indicators. The correlation results are subjected to statistical tests such as t-tests or p-values ​​to evaluate their significance. Based on the significance results and correlation results, the correlation between the fault type and battery performance is judged. For example, if the fault occurs in the charging circuit of the battery, it can be found that there is a negative correlation with the charging efficiency and temperature increase, indicating that the reduced charging efficiency may be due to poor battery connection or unstable current.

[0143] Specifically, the performance difference of each battery is calculated by weighting the health status, charging efficiency, internal resistance and other indicators of different batteries, combining the correlation results and comparison results. For example, performance difference = f(fault correlation result, battery health status, battery temperature, internal resistance, etc.).

[0144] In the embodiment of the present invention, a weighting coefficient can be assigned to each battery according to the performance difference of each battery, and the weighting coefficients are used to generate balance control parameters to control the load distribution, charge and discharge cycle, maximum power output, etc. of the battery. The size of the weighting coefficient can be calculated according to the capacity, charge and discharge efficiency, internal resistance and SOH of the battery.

[0145] In an embodiment of the present invention, when the automatic adjustment module performs the function of adjusting the balance control parameters of the marine lithium battery starting power supply to obtain a dynamic balance management strategy, it is specifically used to:

[0146] generating a preliminary update scheme according to the balance control parameters;

[0147] Performing fitness function evaluation on the preliminary updating scheme to obtain a fitness value;

[0148] The initial update scheme is iterated based on the fitness value to obtain a dynamic balance management strategy.

[0149] In detail, according to the basic parameters of the battery (such as voltage, current, temperature, etc.), a preliminary battery allocation plan is generated through certain rules (such as minimizing the imbalance between batteries, maximizing the utilization of batteries, etc.). The plan can be a weighted calculation result based on capacity, life or other performance parameters.

[0150] In detail, a fitness value is calculated by evaluating the updated battery allocation scheme using a preset fitness function. The fitness function evaluates the updated battery allocation scheme based on the battery's load balancing, charging and discharging efficiency, fault tolerance, and extended battery life to reflect the effectiveness and advantages and disadvantages of the current allocation scheme.

[0151] In detail, the battery allocation scheme is iteratively updated using an optimization algorithm (such as genetic algorithm, particle swarm optimization, simulated annealing, etc.) using the fitness value as a guide. After each iteration, a new fitness value is calculated and the allocation strategy is adjusted based on the result until a preset equilibrium state is reached or the maximum fitness value is reached.

[0152] In an embodiment of the present invention, when the automatic adjustment module applies the dynamic balancing management strategy to the marine lithium battery starting power supply BMS to realize the function of dynamic management of the marine lithium battery starting power supply, it is specifically used to:

[0153] Transmit the dynamic balancing management strategy to the BMS system through the communication interface;

[0154] The BMS system receives the dynamic balancing management strategy and parses the balancing control parameters and control instructions in the dynamic balancing management strategy;

[0155] Based on the control instruction, the control parameters in the BMS system are adjusted according to the balance control parameters in the dynamic balance management strategy.

[0156] In the embodiment of the present invention, various parameters and instructions in the dynamic balancing management strategy are packaged into data frames to ensure the integrity and accuracy of the data, and the data frames are transmitted to the BMS system through the selected communication interface to ensure the reliability and real-time performance of data transmission and avoid data loss or delay.

[0157] In the embodiment of the present invention, the BMS system receives data frames from the main control system through the communication interface, parses the received data frames, extracts the balance control parameters and control instructions in the dynamic balance management strategy, and parses the extracted parameters and instructions into a format that can be understood and executed by the BMS system.

[0158] In the embodiment of the present invention, the corresponding control parameters in the BMS system are adjusted according to the analyzed balance control parameters. This may include: Current regulation: adjusting the charge and discharge current to ensure the current balance of each battery cell. Voltage regulation: adjusting the voltage of each battery cell to prevent overcharging or over-discharging. Temperature control: maintaining the temperature of each battery cell within a reasonable range through heat dissipation or heating measures. Charge and discharge management: optimizing the charge and discharge process to extend the battery life.

[0159] In the embodiment of the present invention, the generated dynamic balancing management strategy is applied to the marine lithium battery starting power supply BMS to achieve dynamic management of the marine lithium battery starting power supply.

[0160] like Figure 4 FIG. 1 is a flow chart of a BMS protection system for a marine lithium battery starting power supply provided by an embodiment of the present invention. In an embodiment of the present invention, the BMS protection system for a marine lithium battery starting power supply includes:

[0161] S401, collecting key parameter data of a marine lithium battery starting power supply, performing feature extraction on the key parameter data, and obtaining extracted features;

[0162] S402, calculating the battery status of the marine lithium battery starting power supply using the extracted features;

[0163] S403, using the battery status and the extracted features to perform potential fault detection on the marine lithium battery starting power supply to obtain a fault location;

[0164] S404: Generate a dynamic balancing management strategy for the marine lithium battery starting power supply according to the fault location, apply the dynamic balancing management strategy to the marine lithium battery starting power supply BMS, and realize dynamic management of the marine lithium battery starting power supply.

[0165] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology.

[0166] Among them, Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0167] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or systems stated in the system can also be implemented by one unit or system through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0168] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A marine lithium battery starting power supply BMS protection system, characterized in that: The system includes: a data extraction module, a state calculation module, a fault detection module, and an automatic adjustment module. Specifically: A data extraction module is used to collect key parameter data of a marine lithium battery starting power supply, and perform data filtering on the key parameter data to obtain a filtering result; Standardizing the filtering result to obtain standardized data; The standardized data is subjected to Fourier transform to obtain frequency domain features, and Fourier transform is performed using the following formula: ； in, is the frequency domain feature frequency components, For the Standardized data, is an imaginary unit, is the total number of standardized data, is the frequency index, ranging from 0 to , is the preset value of pi, is the base of the preset natural logarithm; Calculate the mutual information value between each frequency domain feature and the preset target variable, and use the following formula to calculate the mutual information value: ； in, is the mutual information value, For the frequency domain features, is the target variable, For the Frequency domain features The value of And the target variable value is The probability of For the Frequency domain features The value of Probability, The value of the target variable is The probability of For the Frequency domain features A specific value of Represents the target variable A specific value of ; Sorting the frequency domain features using the mutual information values ​​to obtain a sorting result; According to the sorting results, a preset number of frequency domain features are sequentially selected as extraction features; a state calculation module, configured to reshape the extracted features to obtain a three-dimensional array, sequentially perform forget gate calculations on the three-dimensional array to obtain forget outputs, perform input gate calculations on the three-dimensional array to obtain input values, calculate candidate information according to the input values, calculate cell states according to the candidate information, the input values ​​and the forget outputs, calculate output information according to the cell states, and calculate battery states according to the output information and the cell states; A fault detection module, used to perform potential fault detection on the marine lithium battery starting power supply by using the battery status and the extracted features to obtain a fault location; The automatic adjustment module is used to calculate the performance difference between batteries according to the fault location and the battery status, generate a balance control parameter according to the performance difference, generate a preliminary update plan according to the balance control parameter, perform fitness function evaluation on the preliminary update plan to obtain a fitness value, perform allocation plan update iteration on the preliminary update plan according to the fitness value to obtain a dynamic balance management strategy, apply the dynamic balance management strategy to the marine lithium battery starting power supply BMS, and realize dynamic management of the marine lithium battery starting power supply.

2. A marine lithium battery starting power supply BMS protection system as claimed in claim 1, characterized in that: When the state calculation module performs the function of calculating the cell state according to the candidate information, the input value and the forgotten output, it is specifically used to: Multiply the input value and the candidate information element by element to obtain current information; Adding the forgotten output and the current information to obtain a state value of the cell to be processed; Performing numerical mapping on the state value of the cell to be processed to obtain the cell state.

3. A marine lithium battery starting power supply BMS protection system as claimed in claim 1, characterized in that: When the fault detection module performs the function of performing potential fault detection on the marine lithium battery starting power supply by using the battery status and the extracted features to obtain the fault location, it is specifically used to: Perform feature fusion on the battery status and the extracted features, Get the feature vector; Performing category confidence calculation on the feature vector to obtain confidence; converting the confidence into a probability distribution; The type and location of the potential fault are determined according to the probability distribution.

4. A marine lithium battery starting power supply BMS protection system as claimed in claim 1, characterized in that: When the automatic adjustment module performs the function of generating a dynamic balancing management strategy for the marine lithium battery starting power supply according to the fault location, it is specifically used to: Calculating the performance difference between batteries according to the fault location and the battery status; generating a balance control parameter according to the performance difference; The balance control parameters of the marine lithium battery starting power supply are adjusted to obtain a dynamic balance management strategy.

5. A marine lithium battery starting power supply BMS protection system as claimed in claim 1, characterized in that: When the automatic adjustment module applies the dynamic balancing management strategy to the marine lithium battery starting power supply BMS to realize the function of dynamic management of the marine lithium battery starting power supply, it is specifically used to: Transmit the dynamic balancing management strategy to the BMS system through the communication interface; The BMS system receives the dynamic balancing management strategy and parses the balancing control parameters and control instructions in the dynamic balancing management strategy; Based on the control instruction, the control parameters in the BMS system are adjusted according to the balance control parameters in the dynamic balance management strategy.

Images